#include "clustercommand.h"
+//**********************************************************************************************************************
+vector<string> ClusterCommand::getValidParameters(){
+ try {
+ string AlignArray[] = {"cutoff","precision","method","showabund","timing","hard","outputdir","inputdir"};
+ vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ClusterCommand", "getValidParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+ClusterCommand::ClusterCommand(){
+ try {
+ abort = true;
+ //initialize outputTypes
+ vector<string> tempOutNames;
+ outputTypes["list"] = tempOutNames;
+ outputTypes["rabund"] = tempOutNames;
+ outputTypes["sabund"] = tempOutNames;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ClusterCommand", "ClusterCommand");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+vector<string> ClusterCommand::getRequiredParameters(){
+ try {
+ vector<string> myArray;
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ClusterCommand", "getRequiredParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+vector<string> ClusterCommand::getRequiredFiles(){
+ try {
+ string Array[] = {"phylip","column","or"};
+ vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ClusterCommand", "getRequiredFiles");
+ exit(1);
+ }
+}
//**********************************************************************************************************************
//This function checks to make sure the cluster command has no errors and then clusters based on the method chosen.
ClusterCommand::ClusterCommand(string option) {
else {
//valid paramters for this command
- string Array[] = {"cutoff","precision","method","showabund","timing","outputdir","inputdir"};
+ string Array[] = {"cutoff","precision","method","showabund","timing","hard","outputdir","inputdir"};
vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
OptionParser parser(option);
}
}
+ //initialize outputTypes
+ vector<string> tempOutNames;
+ outputTypes["list"] = tempOutNames;
+ outputTypes["rabund"] = tempOutNames;
+ outputTypes["sabund"] = tempOutNames;
+
//if the user changes the output directory command factory will send this info to us in the output parameter
outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; }
length = temp.length();
convert(temp, precision);
+ temp = validParameter.validFile(parameters, "hard", false); if (temp == "not found") { temp = "F"; }
+ hard = m->isTrue(temp);
+
temp = validParameter.validFile(parameters, "cutoff", false);
if (temp == "not found") { temp = "10"; }
convert(temp, cutoff);
- cutoff += (5 / (precision * 10.0));
+ cutoff += (5 / (precision * 10.0));
method = validParameter.validFile(parameters, "method", false);
if (method == "not found") { method = "furthest"; }
- if ((method == "furthest") || (method == "nearest") || (method == "average")) { }
- else { m->mothurOut("Not a valid clustering method. Valid clustering algorithms are furthest, nearest or average."); m->mothurOutEndLine(); abort = true; }
+ if ((method == "furthest") || (method == "nearest") || (method == "average") || (method == "weighted")) { }
+ else { m->mothurOut("Not a valid clustering method. Valid clustering algorithms are furthest, nearest, average, and weighted."); m->mothurOutEndLine(); abort = true; }
showabund = validParameter.validFile(parameters, "showabund", false);
if (showabund == "not found") { showabund = "T"; }
if (method == "furthest") { cluster = new CompleteLinkage(rabund, list, matrix, cutoff, method); }
else if(method == "nearest"){ cluster = new SingleLinkage(rabund, list, matrix, cutoff, method); }
else if(method == "average"){ cluster = new AverageLinkage(rabund, list, matrix, cutoff, method); }
+ else if(method == "weighted"){ cluster = new WeightedLinkage(rabund, list, matrix, cutoff, method); }
tag = cluster->getTag();
- if (outputDir == "") { outputDir += hasPath(globaldata->inputFileName); }
- fileroot = outputDir + getRootName(getSimpleName(globaldata->inputFileName));
+ if (outputDir == "") { outputDir += m->hasPath(globaldata->inputFileName); }
+ fileroot = outputDir + m->getRootName(m->getSimpleName(globaldata->inputFileName));
- openOutputFile(fileroot+ tag + ".sabund", sabundFile);
- openOutputFile(fileroot+ tag + ".rabund", rabundFile);
- openOutputFile(fileroot+ tag + ".list", listFile);
+ m->openOutputFile(fileroot+ tag + ".sabund", sabundFile);
+ m->openOutputFile(fileroot+ tag + ".rabund", rabundFile);
+ m->openOutputFile(fileroot+ tag + ".list", listFile);
- outputNames.push_back(fileroot+ tag + ".sabund");
- outputNames.push_back(fileroot+ tag + ".rabund");
- outputNames.push_back(fileroot+ tag + ".list");
+ outputNames.push_back(fileroot+ tag + ".sabund"); outputTypes["sabund"].push_back(fileroot+ tag + ".sabund");
+ outputNames.push_back(fileroot+ tag + ".rabund"); outputTypes["rabund"].push_back(fileroot+ tag + ".rabund");
+ outputNames.push_back(fileroot+ tag + ".list"); outputTypes["list"].push_back(fileroot+ tag + ".list");
}
}
}
void ClusterCommand::help(){
try {
m->mothurOut("The cluster command can only be executed after a successful read.dist command.\n");
- m->mothurOut("The cluster command parameter options are method, cuttoff, precision, showabund and timing. No parameters are required.\n");
+ m->mothurOut("The cluster command parameter options are method, cuttoff, hard, precision, showabund and timing. No parameters are required.\n");
m->mothurOut("The cluster command should be in the following format: \n");
m->mothurOut("cluster(method=yourMethod, cutoff=yourCutoff, precision=yourPrecision) \n");
m->mothurOut("The acceptable cluster methods are furthest, nearest and average. If no method is provided then furthest is assumed.\n\n");
if (globaldata->getFormat() == "phylip") { globaldata->setPhylipFile(""); }
else if (globaldata->getFormat() == "column") { globaldata->setColumnFile(""); }
sabundFile.close();rabundFile.close();listFile.close();
- for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); }
+ for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear();
return 0;
}
- if (print_start && isTrue(timing)) {
+ if (print_start && m->isTrue(timing)) {
m->mothurOut("Clustering (" + tag + ") dist " + toString(matrix->getSmallDist()) + "/"
- + toString(roundDist(matrix->getSmallDist(), precision))
+ + toString(m->roundDist(matrix->getSmallDist(), precision))
+ "\t(precision: " + toString(precision) + ", Nodes: " + toString(matrix->getNNodes()) + ")");
cout.flush();
print_start = false;
loops++;
cluster->update(cutoff);
+
float dist = matrix->getSmallDist();
- float rndDist = roundDist(dist, precision);
+ float rndDist;
+ if (hard) {
+ rndDist = m->ceilDist(dist, precision);
+ }else{
+ rndDist = m->roundDist(dist, precision);
+ }
if(previousDist <= 0.0000 && dist != previousDist){
printData("unique");
oldList = *list;
}
- if (print_start && isTrue(timing)) {
+ if (print_start && m->isTrue(timing)) {
m->mothurOut("Clustering (" + tag + ") for distance " + toString(previousDist) + "/" + toString(rndPreviousDist)
+ "\t(precision: " + toString(precision) + ", Nodes: " + toString(matrix->getNNodes()) + ")");
cout.flush();
//delete globaldata's copy of the sparsematrix and listvector to free up memory
delete globaldata->gSparseMatrix; globaldata->gSparseMatrix = NULL;
delete globaldata->gListVector; globaldata->gListVector = NULL;
-
+
//saves .list file so you can do the collect, rarefaction and summary commands without doing a read.list
if (globaldata->getFormat() == "phylip") { globaldata->setPhylipFile(""); }
else if (globaldata->getFormat() == "column") { globaldata->setColumnFile(""); }
sabundFile.close();
rabundFile.close();
listFile.close();
-
- if (saveCutoff != cutoff) { m->mothurOut("changed cutoff to " + toString(cutoff)); m->mothurOutEndLine(); }
+
+ if (saveCutoff != cutoff) {
+ if (hard) { saveCutoff = m->ceilDist(saveCutoff, precision); }
+ else { saveCutoff = m->roundDist(saveCutoff, precision); }
+
+ m->mothurOut("changed cutoff to " + toString(cutoff)); m->mothurOutEndLine();
+ }
m->mothurOutEndLine();
m->mothurOut("Output File Names: "); m->mothurOutEndLine();
m->mothurOutEndLine();
- //if (isTrue(timing)) {
+ //if (m->isTrue(timing)) {
m->mothurOut("It took " + toString(time(NULL) - estart) + " seconds to cluster"); m->mothurOutEndLine();
//}
void ClusterCommand::printData(string label){
try {
- if (isTrue(timing)) {
+ if (m->isTrue(timing)) {
m->mothurOut("\tTime: " + toString(time(NULL) - start) + "\tsecs for " + toString(oldRAbund.getNumBins())
+ "\tclusters. Updates: " + toString(loops)); m->mothurOutEndLine();
}
start = time(NULL);
oldRAbund.setLabel(label);
- if (isTrue(showabund)) {
+ if (m->isTrue(showabund)) {
oldRAbund.getSAbundVector().print(cout);
}
oldRAbund.print(rabundFile);